A position estimation system including a first arrangement for providing an image with a known target in a known reference frame. A second arrangement correlates the image with a stored image. The correlation is used to compute an error with respect to a position estimate. In a specific embodiment, the error is referenced with respect to first (x), second (y) and third (z) directions. A target location error is computed with respect to a stored image provided by a target image catalog. The target image catalog includes target geo-locations and digital terrain elevation data. In an illustrative application, the image data is provided by synthetic aperture radar and forward-looking infrared systems. An observation model and a measure noise matrix are Kalman filtered to ascertain a position error in navigation data generated by an integrated inertial navigation and global positioning system.
|
21. A position estimation method including the steps of:
providing an image including a known target in a known reference frame;
correlating said image with a stored image;
computing an error in response said step of correlating said image with a stored image, wherein said error is referenced with respect to first (x), second (y) and third (z) directions;
providing an observation model in, response to said error; and
filtering said model to provide a position estimate.
1. A position estimation system comprising:
first means for providing an image including a known target in a known reference frame;
second means for correlating said image with a stored image;
third means responsive to said second means for computing an error in response thereto, wherein said error is referenced with respect to first (x), second (y) and third (z) directions;
fourth means responsive to said third means for providing an observation model; and
fifth means for filtering said model to provide a position estimate.
40. A position estimation system comprising:
first means for providing an image including a known target in a known reference frame;
second means for correlating said image with a stored image;
third means responsive to said second means for computing a position estimate error in response thereto, wherein said error is referenced with respect to first (x), second (y) and third (z) directions;
fourth means for providing an observation matrix;
fifth means for providing a measurement noise matrix; and
sixth means for using an observation matrix and a measurement noise matrix to correct for said error.
39. A navigation method including the steps of:
providing an inertial navigation system;
providing a global positioning system;
minimizing an error in position data generated by said inertial navigation system in response to an output from said receiver, wherein said error is referenced with respect to first (x), second (y) and third (z) directions;
detecting interference in reception of said receiver and providing a signal in response thereto;
referencing data from a target in a known location to minimize an error generated by said inertial navigation system in response to said signal;
providing an observation model in response to said error; and
filtering said model to provide a position estimate.
20. A navigation system comprising:
an inertial navigation system;
a global positioning system receiver for minimizing an error generated by said inertial navigation system;
means for minimizing an err or in position data generated by said inertial navigation system in response to an output from said receiver, wherein said error is referenced with respect to first (x), second (y) and third (z) directions;
means for detecting interference in reception of said receiver and providing a signal in response thereto;
responsive to said signal for referencing data from target in a known location to minimize an error generated by said inertial navigation system;
means responsive to said means for minimizing for providing an observation model; and
means for filtering said model to provide a position estimate.
7. The invention of
8. The invention of
9. The invention of
10. The invention of
11. The invention of
12. The invention of
15. The invention of
16. The invention of
19. The invention of
22. The invention of
29. The invention of
30. The invention of
31. The invention of
32. The invention of
33. The invention of
34. The invention of
35. The invention of
38. The invention of
|
1. Field of the Invention
The present invention relates to inertial navigation and global position. More specifically, the present invention relates to systems and methods for improving the GPS counter-jamming performance of inertial navigation systems.
2. Description of the Related Art
Inertial navigation systems typically use gyroscopes and accelerometers to provide precision vehicular navigation. Unfortunately, inertial navigation accuracy degrades because of instrument calibration errors and other errors. These navigation errors typically grow as a function of time. Independent observations of the vehicle navigation information are needed to bind these navigation errors. Therefore, sensors, other than INS, are needed in order to obtain independent navigation information. Hence, a conventional approach for correcting these errors involves the integration of a Global Position System (GPS) receiver with the inertial navigation system. However, the GPS is vulnerable to jamming which can impede the ability of the GPS system to correct the inertial navigation errors.
Typically, to counter the effects of GPS jamming, designers have endeavored to: 1) improve the accuracy of the inertial navigation system and 2) make the GPS receiver resistant to jamming. However, these approaches are expensive and limited in efficacy.
Hence, a need remains in the art for an effective yet inexpensive system or method for improving the navigation accuracy of integrated inertial navigation and Global Positioning Systems.
The need in the art is addressed by the position estimation system of the present invention. In a most general implementation, the inventive system includes a first arrangement for providing an image including a known target in a known reference frame. A second arrangement correlates the image with a stored image. The correlation is used to compute an error with respect to a position estimate.
In a specific embodiment, the error is referenced with respect to first (x), second (y) and third (z) directions. A target location error is computed with respect to a stored image provided by a target image catalog. The target image catalog includes target geo-locations and digital terrain elevation data. In an illustrative application, the image data is provided by synthetic aperture radar or forward-looking infrared systems. An observation model and a measure noise matrix are Kalman filtered to ascertain a position error in navigation data generated by an integrated inertial navigation and Global Positioning system.
In the illustrative application, geo-registered SAR/FLIR imagery is used to track targets and to determine a target location error (TLE). This TLE information is a set of error equations that describe the relationship between vehicle navigation information and target data. In accordance with the invention, this relationship is used to form an observation model for vehicle navigation with respect to target locations. Using Kalman filtering and the observation model, vehicle navigation errors can be bound and the navigation accuracy of the vehicle can be improved.
Illustrative embodiments and exemplary applications will now be described with reference to the accompanying drawings to disclose the advantageous teachings of the present invention.
While the present invention is described herein with reference to illustrative embodiments for particular applications, it should be understood that the invention is not limited thereto. Those having ordinary skill in the art and access to the teachings provided herein will recognize additional modifications, applications, and embodiments within the scope thereof and additional fields in which the present invention would be of significant utility.
In general, in accordance with the present teachings, geo-registered SAR/FLIR imagery is used to track targets and to determine a target location error (TLE). The TLE is generated by a set of error equations that describe the relationship between the vehicle (sensor) navigation information and the target location. The geo-registered images obtained from SAR/FLIR systems provide position estimates for a known pixel on the ground or other reference frame through target recognition methods. This position estimate serves as an independent observation to bind errors in an inertial navigation system.
U.S. Pat. No. 5,485,384 entitled ON-BOARD NAVIGATION SYSTEM FOR AN AERIAL CRAFT INCLUDING A SYNTHETIC APERTURE SIDEWAYS LOOKING RADAR issued Jan. 16, 1996 to B. Falconnet (hereinafter the “Falconnet” patent) the teachings of which are hereby incorporated herein by reference appears to teach the use of SAR (Synthetic Aperture Radar) sensors to obtain target imagery in the x and y horizontal plane. (See also U.S. Pat. No. 5,432,520 issued Jul. 11, 1995 to Schneider et al. and entitled SAR/GPS INERTIAL METHOD OF RANGE MEASUREMENT, the teachings of which are herby incorporated herein by reference.) This imagery is then correlated with maps of the geo-locations that are pre-stored in the database to obtain two error equations in the x and y directions. These two error equations serve as an observation model for the Kalman filter to bind the vehicle navigation errors.
In accordance with the present invention, Falconnet's teachings are extended by: 1) including a third dimensional axis, the altitude of a target image location and 2) providing a specific teaching as to how the third dimension can be used to improve the navigational accuracy of an integrated INS/GPS navigation system. The geo-registered imagery is extended to sensors from SAR or FLIR (forward-looking infrared) systems. A simple first order error model in the computed target geo-location is used to illustrate the effectiveness of Kalman filter updating using geo-registration imagery. Detailed x, y, and z observation equations are provided which involve the vehicle's position, velocity, and attitude, as well as the angle between the horizontal plane and the slant plane. The position error differences can be minimized through optimal estimation techniques, such as Kalman filter, to bind INS navigation errors. The equations form an observation matrix in a Kalman filter.
The method described in this invention can also be extended to any sensor on the vehicle that produces target location errors (TLE) on the known target image because the TLE equations can be reduced to a set of equations related to the-vehicle navigation errors and target image position errors.
If J<T, then the data from the INS and GPS processors 22 and 26 is input to the Kalman filter 28 (
If, however, J≧T, then at step 108 the system 10 generates target location error (TLE) equations drawing data from the catalog of geo-registered features 18 (
Returning to
Errors in computed target geo-location can be primarily attributed to three major sources: sensor position errors, sensor bearing errors and DTED errors. In accordance with the present teachings, these errors are treated as being statistically independent and zero mean Gaussian. For the sake of simplicity, all other errors are assumed to be relatively small and are ignored. It is also assumed that the image of the target can be acquired. The DTED noise is treated herein as part of the measurement noise in the z-direction of an ECEF (Earth Center Earth Fixed) coordinate frame.
Errors in the sensor are directly transformed into the slant coordinate frame and then transformed to the platform coordinate frame. The slant coordinate frame is defined as follows: xs is along the vehicle velocity axis {right arrow over (V)}, the zs is perpendicular to the slant plane which is the plane that passes through {right arrow over (V)}D and {right arrow over (R)}, and ys forms a right-hand coordinate system.
Therefore, errors due to the sensor position errors in the slant coordinate frame are derived as follows:
where Δ{right arrow over (r)} and Δ{right arrow over (V)} are the vectors of the vehicle navigation position and velocity errors, respectively, in the body coordinate frame. {right arrow over (V)} is the velocity vector of the vehicle, {right arrow over (R)} is the range vector from the vehicle position to the known geo-location, and φ is the angle between the vehicle flight path and line of sight between vehicle and the target image.
The errors due to sensor bearing errors, in the slant coordinate frame, are derived as follows:
Next, we combine the errors provided by equations [1] and [2] to obtain the following errors dxs, dys, and dzs that are in the slant coordinate frame.
Next, converting these errors, dxs, dys, and dzs into the platform coordinate frame by the angle ψ (see
Inserting equation [3] into equation [4]:
where:
The classical Kalman filter is described as follows:
{right arrow over (x)}k=Φk-1{right arrow over (x)}k-1+{right arrow over (w)}k-1, {right arrow over (w)}k˜N({right arrow over (0)},Qk)
{right arrow over (z)}k=Hk{right arrow over (x)}k+{right arrow over (v)}k, {right arrow over (v)}k˜N({right arrow over (0)},Rk)
E[{right arrow over (x)}(0)]={right arrow over (x)}0, E[{right arrow over (x)}0{right arrow over (x)}0T]=P0, E[{right arrow over (w)}j{right arrow over (v)}kT]={right arrow over (0)}∀j,k [7]
Pk−=Φk-1Pk-1+Φk-1T+Qk-1
Pk+=(I−KkHk)Pk−
Kk=Pk−HkT[HkPk−HkT+Rk]−1
where Φk is the transition matrix.
In accordance with the present teachings, an observation matrix Hk and the observation (measurement) noise Rk are generated as follows. Assume that Δ{right arrow over (r)} and Δ{right arrow over (V)} are the first six Kalman filter error states defined as Δ{right arrow over (r)}=(δrx, δry, δrz) T and Δ{right arrow over (V)}=(δVx, δVy, δVz)T in the platform coordinate frame where superscript T denotes the transpose of the vector. Note that if Δ{right arrow over (r)} and Δ{right arrow over (V)} are defined in the ECEF coordinate frame, then these error states need to be transformed into the platform frame where the error equations [5] are defined. The Kalman filter error state {right arrow over (x)}k, observation matrix Hk, and measurement noise matrix Rk are denote as below:
Assume that the vectors {right arrow over (R)}=(Rx, Ry, Rz), V=|{right arrow over (V)}| (magnitude of {right arrow over (V)}), R=|{right arrow over (R)}|,
Therefore, equation [5] can be expressed in the following forms:
Therefore, the elements in the observation matrix are as follows:
h00=px+Rx·sin φ
h01=py+Ry·sin φ
h02=pz+Rz·sin φ
h03=h00/V
h04=h01/V
h10=Rx·sin φcos ψ−qx·sin ψ
h11=Ry·sin φcos ψ−qy·sin ψ
h12=Rz·sin φcos ψ−qz·sin ψ
h13=h10/V [10]
h14=h11/V
h15=h12/V
h20=Rx·sin φsin ψ+qx·cos ψ
h21=Ry·sin φsin ψ+qy·cos ψ
h22=Rz·sin φsin ψ+qz·cos ψ
h23=h20/V
h24=h21/V
h25=h22/V
and r00, r11, and r22 represent the observation noise in the SAR/FLIR imager accuracy during the geo-target scan. The DTED noise is included in r22 term.
In the illustrative embodiment, the sensor position and attitude errors are converted into the slant coordinate frame and then into the platform frame. Errors that may be due to the SAR signal processing and ranging errors are ignored since these errors are assumed to be relatively small. If some of those errors are large enough to be considered, the same method can be used to transform these errors into the platform frame and obtain similar results as equation [11] below:
The error states in the Kalman filter should include the vehicle navigation errors such as {right arrow over (x)}k=(Δrx, Δry, Δrz, ΔVx, ΔVy, ΔVz, . . . )T. The error equations above can be converted to an observation matrix in the Kalman filer as follows:
where hij are defined above.
In general, coordinate frames of the sensor errors and navigation error states in the Kalman filter are different. They need to transform into the same coordinate frame.
The error equations in equation [1] are, in fact, a TLE in the platform coordinate frame. Based on this invention, any TLE can be extracted and converted into an observation matrix as described in equation [5].
If the sensor SAR/FLIR can only obtain two-dimensional images (dxp, dyp), the observation matrix will be a two-dimensional matrix:
where the elements hij are defined as before.
Thus, the present invention has been described herein with reference to a particular embodiment for a particular application. Those having ordinary skill in the art and access to the present teachings will recognize additional modifications applications and embodiments within the scope thereof.
It is therefore intended by the appended claims to cover any and all such applications, modifications and embodiments within the scope of the present invention.
Accordingly,
Chiou, Kuo-Liang, Chiou, Carroll C., Rudolph, Kevin E.
Patent | Priority | Assignee | Title |
10444347, | Nov 30 2016 | GM Global Technology Operations LLC | Accurate self localization using automotive radar synthetic aperture radar |
10663579, | Jan 27 2017 | Massachusetts Institute of Technology | Extending footprint for localization using surface penetrating radar (SPR) |
10725171, | Jan 27 2017 | Massachusetts Institute of Technology | Method and system for localization of a vehicle using surface penetrating radar |
10746867, | Jan 27 2017 | Massachusetts Institute of Technology | Determining surface characteristics |
10802135, | Dec 21 2016 | The Boeing Company | Method and apparatus for raw sensor image enhancement through georegistration |
10809064, | Feb 08 2018 | Raytheon Company | Image geo-registration for absolute navigation aiding using uncertainy information from the on-board navigation system |
11175398, | Dec 21 2016 | The Boeing Company | Method and apparatus for multiple raw sensor image enhancement through georegistration |
11402493, | Jan 27 2017 | Massachusetts Institute of Technology | Determining surface characteristics |
8155433, | Jul 10 2008 | GOODRICH CORPORATION | Method of object location in airborne imagery using recursive quad space image processing |
8373127, | Jun 26 2008 | Lynntech, Inc | Method of searching for a thermal target |
8406513, | Jul 10 2008 | GOODRICH CORPORATION | Method of object location in airborne imagery using recursive quad space image processing |
8548194, | Nov 12 2008 | L3 Technologies, Inc | System and method for determining altitude |
8949024, | Oct 25 2012 | Massachusetts Institute of Technology | Vehicle localization using surface penetrating radar |
Patent | Priority | Assignee | Title |
4495580, | Mar 30 1981 | E-Systems, Inc. | Navigation system |
5432520, | Oct 18 1993 | HE HOLDINGS, INC , A DELAWARE CORP | SAR/GPS inertial method of range measurement |
5485384, | Sep 03 1992 | AEROSPATIALE SOCIETE NATIONALE INDUSTRIELLE AYANT SON SIEGE SOCIAL | On-board navigation system for an aerial craft including a synthetic aperture sideways looking radar |
5963653, | Jun 19 1997 | Raytheon Company | Hierarchical information fusion object recognition system and method |
6496778, | Sep 14 2000 | American GNC Corporation | Real-time integrated vehicle positioning method and system with differential GPS |
6630904, | Feb 02 1999 | CHARLES STARK DRAPER LABORATORY, INC , THE | Deeply-integrated adaptive INS/GPS navigator with extended-range code tracking |
6697736, | Feb 06 2002 | American GNC Corporation | Positioning and navigation method and system thereof |
6744401, | May 13 2002 | Honeywell International Inc | Methods and apparatus for radar data processing |
6865477, | May 31 1994 | B E AEROSPACE, INC | High resolution autonomous precision positioning system |
6912464, | Jul 14 1997 | BAE SYSTEMS, plc | Inertial navigation accuracy enhancement |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
Jun 21 2005 | CHIOU, KUO-LIANG | Raytheon Company | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 016726 | /0813 | |
Jun 21 2005 | RUDOLPH, KEVIN E | Raytheon Company | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 016726 | /0813 | |
Jun 21 2005 | CHIOU, CARROLL C | Raytheon Company | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 016726 | /0813 | |
Jun 23 2005 | Raytheon Company | (assignment on the face of the patent) | / |
Date | Maintenance Fee Events |
Sep 21 2011 | M1551: Payment of Maintenance Fee, 4th Year, Large Entity. |
Dec 16 2015 | M1552: Payment of Maintenance Fee, 8th Year, Large Entity. |
Dec 19 2019 | M1553: Payment of Maintenance Fee, 12th Year, Large Entity. |
Date | Maintenance Schedule |
Jul 01 2011 | 4 years fee payment window open |
Jan 01 2012 | 6 months grace period start (w surcharge) |
Jul 01 2012 | patent expiry (for year 4) |
Jul 01 2014 | 2 years to revive unintentionally abandoned end. (for year 4) |
Jul 01 2015 | 8 years fee payment window open |
Jan 01 2016 | 6 months grace period start (w surcharge) |
Jul 01 2016 | patent expiry (for year 8) |
Jul 01 2018 | 2 years to revive unintentionally abandoned end. (for year 8) |
Jul 01 2019 | 12 years fee payment window open |
Jan 01 2020 | 6 months grace period start (w surcharge) |
Jul 01 2020 | patent expiry (for year 12) |
Jul 01 2022 | 2 years to revive unintentionally abandoned end. (for year 12) |